Long-term Effects of High and Low Glycemic Load Diets at Different Levels of Caloric Restriction on Dietary Adherence, Body Composition and Metabolism in CALERIE, a One Year Randomized Controlled Trial

Authors: DAS, SAI KRUPA - HNRCA AT TUFTS; GILHOOLY, CHERYL - HNRCA AT TUFTS; GOLDEN, JULIE - HNRCA AT TUFTS & FDA; PITTAS, ANASTASSIOS - TUFTS-NEW ENGLAND MEDICAL; FUSS, PAUL - HNRCA AT TUFTS; CHEATAM, RACHEL - HNRCA AT TUFTS; TYLER, STEPHANIE - HNRCA AT TUFTS; TSAY, MICHELLE - HNRCA AT TUFTS; MCCRORY, MEGAN - BASTYR UNIVERSITY; Lichtenstein, Alice; Dallal, Gerald; DUTTA, CHANDA - NIA, NIH; BHAPKAR, MANJUSHRI - DUKE CLINICAL RESEARCH; DELANEY, JAMES - UNIV OF PITTSBURGH MED.; Saltzman, Edward; Roberts, Susan

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/13/2006
Publication Date: 4/1/2007
Citation: Das, S., Gilhooly, C.H., Golden, J.K., Pittas, A.G., Fuss, P.J., Cheatam, R., Tyler, S., Tsay, M., Mccrory, M.A., Lichtenstein, A.H., Dallal, G., Dutta, C., Bhapkar, M.V., Delaney, J.P., Saltzman, E., Roberts, S. 2007. Long-term Effects of High and Low Glycemic Load Diets at Different Levels of Caloric Restriction on Dietary Adherence, Body Composition and Metabolism in CALERIE, a One Year Randomized Controlled Trial. American Journal of Clinical Nutrition. 85(4):1023-1030.

Interpretive Summary: The prevalence of obesity and overweight continue to increase nationally and worldwide Caloric restriction (CR) remains the cornerstone of most weight management strategies, but there remains no consensus over the precise roles of dietary macronutrient composition and level of caloric deficit in optimizing long-term weight loss. In part, the lack of consensus probably reflects the fact that most studies in this area have provided dietary advice, rather than food, with resulting uncertainty in the true extent of dietary change. Further unbiased assessments of adherence to the tested regimens were not performed or differential behavioral support was given to each treatment group because the studies were testing popular diet prescriptions rather than specifically different dietary compositions, which confounded the result. The question of what level of dietary energy restriction is most effective for long-term weight loss is probably also important, but has received much less attention than dietary macronutrient composition. It is not known whether diets with higher CR that cause more rapid weight loss initially also have the greatest long-term success, since psychological factors including an individuals’ weight loss goal and the impact of setting attainable vs. unattainable goals may be important. We conducted a detailed 1year randomized clinical trial in healthy overweight women and men (mean+/-SD, age 35+/-6y; BMI 27.6+/-1.4 kg/m2) and were the first to examine the effects of HG vs. LG diets and low (10%) vs moderate (30%) levels of CR on weight loss in a provided-food protocol not confounded by group differences in other factors that strongly influence energy intake including type of behavioral support and palatability and dietary variety of the regimens. Under the conditions of this study, we found similar mean energy intakes, weight loss and body fat loss throughout the 12 month study in all the groups. These findings provide rigorous support for the view that glycemic load has little effect on long–term weight loss during CR. Our findings further indicate that lower levels of CR facilitate similar long-term weight loss and at the same time may be easier to tolerate based on the finding of increased satisfaction with the amount of provided food. Whereas the subjects randomized to 30% CR were unable to meet the study goals for reduced dietary intake as assessed by objective doubly labeled water measurements, subjects randomized to 10% CR frequently exceeded what was asked them and returned some of the provided food. These findings suggest that small reductions in energy intake may be the most effective way to promote long term weight loss, because they allow individuals to meet goals and exceed expectations without compromising long-term weight loss. These results suggest that a broad range of healthful dietary compositions can successfully promote weight loss and that a focus on maintaining low level CR over time rather than one particular macronutrient composition may better enhance individual long-term weight loss success and contribute to reducing the ongoing obesity epidemic.

Technical Abstract: Context The effects of dietary macronutrient composition and level of energy intake on adherence to a calorically-restricted diet remain uncertain. Objective To examine the effects of dietary macronutrients, and level of caloric restriction (CR), for 12 months, on adherence to the prescribed regimens, weight loss, body fat loss and related variables. Design and Settings Randomized clinical trial of high glycemic (HG) load (60% carbohydrate, 20% fat, 20% protein) vs. low glycemic (LG) load (40% carbohydrate, 30% fat, 30% protein) diets at two levels of caloric restriction (10% and 30%) in 46 healthy overweight adults (mean+/-SD, age 35+/-6y; BMI 27.6+/-1.4 kg/m2). All food was provided for the first 6 months in diets controlled for variety, palatability and other variables, and subjects self-administered the dietary prescription for the final 6 months. Consistent behavioral support was offered to all groups. Main outcome measures Changes over time in actual energy intake (measured using the doubly labeled water method), body weight and fatness, hunger and satiety, blood parameters and resting metabolic rate. Results Energy intake measured using doubly labeled water was not significantly different among the groups over time. Both 30% CR groups ate somewhat more food than provided (21.1%CR and 27.5%CR in 30%HG and 30%LG, respectively, at 3 months), while the 10% groups ate less than provided (15.0%CR and 21.5%CR in 10%HG and 10%LG, respectively, at 3 months). Consequently, there were no statistically significant effects of dietary composition or level of CR on weight loss from baseline (6.41+/-3.8 kg in 30% HG vs 6.08+/-4.0. kg in 30%LG; 7.21+/-8.2 in 10% HG vs 2.30+/-5.8 in 10% LG at 12 months). There was also no significant difference between the groups in body fat loss and decrease in resting metabolic rate. Between 6 and 12 mo, when subjects prepared their own food, LG subjects gained body weight and fat and increased RMR compared to HG subjects (P <0.05). Conclusions This high-intensity study in healthy overweight individuals found no difference in actual energy intake, weight or body fat loss over 12 mo between HG and LG diets or between 10% and 30%CR. Combined, these findings suggest that both HG and LG diets can support weight loss and that small long-term reductions in energy intake may be the most effective way to promote long term weight loss.